Sign up to get updates from us
Arthur Conmy sits down with Nathan Labenz for an accessible deep dive into the state of interpretability research online today.
Arthur Conmy sits down with Nathan Labenz for an accessible deep dive into the state of interpretability research online today. They discuss how pioneering researchers have painstakingly worked to isolate the sub-circuits within transformers that are responsible for different aspects of AI capabilities. Arthur also introduces us to a new ACDC approach that he and his co-authors have taken to automating some of the most time-consuming parts of this work. If you’re looking for an ERP platform, check out our sponsor, NetSuite: http://netsuite.com/cognitive
RECOMMENDED PODCAST:
Founding a business is just the tip of the iceberg; the real complexity comes with scaling it. On 1 to 1000, hosts Jack Altman and Erik Torenberg dig deep into the inevitable twists and turns operators encounter along the journey of turning an idea into a business. Hear all about the tactical challenges of scaling from the people that built up the world’s leading companies like Stripe, Ramp, and Lattice. Our first episode with Eric Glyman of Ramp is out now: https://link.chtbl.com/1to1000
RECOMMENDED PODCAST:
Run the Numbers is a weekly podcast about financial metrics and business models, designed for ambitious people operating tech startups. It's a collection of things host CJ Gustafson (CFO at Partstech and writer of Mostly Metrics) has learned and thought about in the trenches as a tech CFO. Subscribe to listen on the platform of your choice: https://link.chtbl.com/runthenumbers
The Cognitive Revolution is a part of the Turpentine podcast network. Learn more: Turpentine.co
TIMESTAMPS:
(00:00) Episode Preview
(04:40) What attracted Arthur to mechanistic interpretability?
(07:49) LLM information processing: General Understanding vs Stochastic Parrot Paradigm
(14:00) ACDC paper: https://arxiv.org/abs/2304.14997
(14:45) Sponsors: NetSuite | Omneky
(24:30) Putting together data sets
(32:39) How to intervene in LLMs network activity
(36:00) Defining metrics to evaluate the production of correct completions
(44:20) The future of the mechanistic interpretability research
(50:00) Extracting upstream activations in the ACDC project and evaluating impact on downstream components.
(56:00) Anthropic research findings
(01:08:00) 3-Step process of the ACDC approach
(01:22:00) Setting a threshold and validation
(01:27:00) Goal of the approach
(01:32:00) Compute requirements
(01:35:30) Scaling laws for mechanistic interpretability
(01:40:00) Accessibility of this research for casual enthusiasts
(01:46:00) Emergence discourse
(01:56:00) Path to AI safety
LINKS:
https://arthurconmy.github.io/
https://arxiv.org/abs/2304.14997
X:
@labenz (Nathan)
@arthurconmy (Arthur)
@cogrev_podcast
SPONSORS: NetSuite | Omneky
-NetSuite provides financial software for all your business needs. More than thirty-six thousand companies have already upgraded to NetSuite, gaining visibility and control over their financials, inventory, HR, eCommerce, and more. If you’re looking for an ERP platform: NetSuite (http://netsuite.com/cognitive) and defer payments of a FULL NetSuite implementation for six months.
-Omneky is an omnichannel creative generation platform that lets you launch hundreds of thousands of ad iterations that *actually work* customized across all platforms, with a click of a button. Omneky combines generative AI and real-time advertising data. Mention "Cog Rev" for 10% off.
